calcBraidBootstrap: BRAID Parameter Confidence Intervals

In braidrm: Fitting Combined Action with the BRAID Response Surface Model

BRAID Parameter Confidence Intervals

Description

Uses residuals-based bootstrapping to estimate confidence intervals on a BRAID fit's response surface parameters

Usage

calcBraidBootstrap(bfit, ciLevs = c(0.025, 0.975), numBoot = NULL)

Arguments

bfit	A BRAID fit object of class braidrm. If this object already has bootstrapped coefficients, a warning will be given, and they will be overwritten
ciLevs	The lower and upper quantiles at which the confidence intervals should be estimated. Default is 0.025 and 0.975, producing 95% confidence intervals
numBoot	The number of bootstrapped coefficient values to estimate. Defaults to a number large enough that at least 10 measurement should lie outside the estimated interval; but held to a minimum value of 100 and a maximum value of 1000.

Value

An object of class braidrm but with three additional elements:

• ciLevs: The two quantiles at which the confidence intervals are set

• ciCoefs: An array of bootstrapped coefficients. The number of rows is the number of successful bootstrapped fits; there is one column for each the nine BRAID parameters

• ciMat: An array of confidence intervals on the fitted parameters. The rows correspond to the free parameters included in the fit, and are named for those parameters; the firt column contains the lower bound of the confidence intervals, the second column the upper bound.

Examples

bfit <- braidrm(measure ~ concA + concB, synergisticExample, getCIs = FALSE)
summary(bfit)

bfit_ci <- calcBraidBootstrap(bfit)
summary(bfit_ci)

braidrm documentation built on Sept. 30, 2024, 9:40 a.m.